From 372f13dc5ad89af45e1c602bc8f0a08a6b428ef2 Mon Sep 17 00:00:00 2001
From: Kim Rutherford <kmr@aska.gen.nz>
Date: Mon, 13 Nov 2023 18:56:06 +1300
Subject: [PATCH] Add a condition_details vec to OntAnnotationDetail

Show we can show the condition details on the website.

Refs pombase/website#2029
---
 src/pombase/data_types.rs     |  4 +++-
 src/pombase/web/data_build.rs | 33 +++++++++++++++++++++++++++++++++
 tests/test_data_build.rs      |  8 ++++++--
 3 files changed, 42 insertions(+), 3 deletions(-)

diff --git a/src/pombase/data_types.rs b/src/pombase/data_types.rs
index a6a8c84f..8476fa48 100644
--- a/src/pombase/data_types.rs
+++ b/src/pombase/data_types.rs
@@ -1,4 +1,4 @@
-use std::collections::{HashMap, HashSet, BTreeMap};
+use std::collections::{HashMap, HashSet, BTreeMap, BTreeSet};
 
 use std::fmt::Display;
 use std::fmt;
@@ -754,6 +754,8 @@ pub struct OntAnnotationDetail {
     pub genotype_background: Option<FlexStr>,
     #[serde(skip_serializing_if="HashSet::is_empty", default)]
     pub conditions: HashSet<TermId>,
+    #[serde(skip_serializing_if="BTreeSet::is_empty", default)]
+    pub condition_details: BTreeSet<(TermId, Option<String>)>,
     #[serde(skip_serializing_if="Option::is_none")]
     pub date: Option<FlexStr>,
     #[serde(skip_serializing_if="Option::is_none")]
diff --git a/src/pombase/web/data_build.rs b/src/pombase/web/data_build.rs
index 3eb8ea0f..acba95cc 100644
--- a/src/pombase/web/data_build.rs
+++ b/src/pombase/web/data_build.rs
@@ -841,6 +841,27 @@ fn set_has_protein_features(genes: &mut UniquenameGeneMap, protein_view_data: &H
     }
 }
 
+
+lazy_static! {
+    static ref CONDITION_DETAIL_RE: Regex =
+        Regex::new(r"^([A-Z][\w]+:\d\d\d+)(?:\((.+)\))?$").unwrap();
+}
+
+// parse something like "FYECO:0000005(32C)" or "FYECO:0000329(2% (v/v))" into a term ID
+// and a String containers the details from the brackets
+// parse "FYECO:0000005" into (TermId, None)
+fn parse_condition_with_detail(condition_string: &str) -> (TermId, Option<String>) {
+    let captures = CONDITION_DETAIL_RE.captures(condition_string).unwrap();
+
+    let term_id = captures.get(1).as_ref().unwrap().as_str().into();
+
+    if let Some(detail) = captures.get(2) {
+        (term_id, Some(detail.as_str().into()))
+    } else {
+        (term_id, None)
+    }
+}
+
 impl <'a> WebDataBuild<'a> {
     pub fn new(raw: &'a Raw,
                domain_data: &'a HashMap<UniprotIdentifier, UniprotResult>,
@@ -3776,6 +3797,7 @@ phenotypes, so just the first part of this extension will be used:
             let publication = &feature_cvterm.publication;
             let mut extra_props: HashMap<FlexStr, FlexStr> = HashMap::new();
             let mut conditions: HashSet<TermId> = HashSet::new();
+            let mut condition_details: BTreeSet<(TermId, Option<String>)> = BTreeSet::new();
             let mut withs: HashSet<WithFromValue> = HashSet::new();
             let mut froms: HashSet<WithFromValue> = HashSet::new();
             let mut qualifiers: Vec<Qualifier> = vec![];
@@ -3838,6 +3860,16 @@ phenotypes, so just the first part of this extension will be used:
                                           value, feature.uniquename, termid);
                             }
                         },
+                    "condition_detail" =>
+                        if let Some(ref value) = prop.value {
+                            if value.contains(':') {
+                                let parsed_value = parse_condition_with_detail(value);
+                                condition_details.insert(parsed_value);
+                            } else {
+                                eprintln!(r#"ignoring condition that doesn't contain a term ID "{}" (from annotation of {} with {})"#,
+                                          value, feature.uniquename, termid);
+                            }
+                        },
                     "qualifier" =>
                         if let Some(value) = prop.value.clone() {
                             qualifiers.push(value);
@@ -3998,6 +4030,7 @@ phenotypes, so just the first part of this extension will be used:
                 evidence,
                 eco_evidence,
                 conditions,
+                condition_details,
                 extension,
                 date,
                 assigned_by,
diff --git a/tests/test_data_build.rs b/tests/test_data_build.rs
index ac701cae..912788d1 100644
--- a/tests/test_data_build.rs
+++ b/tests/test_data_build.rs
@@ -1,4 +1,4 @@
-use std::collections::{HashMap, HashSet};
+use std::collections::{HashMap, HashSet, BTreeSet};
 use std::iter::FromIterator;
 
 use std::cmp::Ordering;
@@ -402,6 +402,9 @@ fn make_one_detail(id: i32, gene_uniquename: &str, reference_uniquename: &str,
                    maybe_genotype_uniquename: Option<&str>, evidence: &str,
                    extension: Vec<ExtPart>,
                    conditions: HashSet<TermId>) -> OntAnnotationDetail {
+    let condition_details: BTreeSet<_> =
+        conditions.iter().map(|cond| (cond.clone(), None)).collect();
+
     OntAnnotationDetail {
         id: id,
         genes: vec![gene_uniquename.into()],
@@ -419,7 +422,8 @@ fn make_one_detail(id: i32, gene_uniquename: &str, reference_uniquename: &str,
         qualifiers: vec![],
         extension: extension,
         gene_ex_props: None,
-        conditions: conditions,
+        conditions,
+        condition_details,
         assigned_by: Some("PomBase".to_shared_str()),
         throughput: Some(Throughput::HighThroughput),
     }